India's AI Regulation Consultation Paper Is Nearly Ready, Reports Say

According to the reports, the paper will deal with AI safety, the spread of deepfakes and the possible psychological effects of AI on users, along with skilling and making sure AI's benefits reach more people. Storyboard18 said the consultation will also look at accountability, copyright, data protection, deepfake detection and institutional arrangements. A senior government official said the ministry had studied regulatory frameworks adopted in other countries and consulted stakeholders, including academics.

The paper is later than first promised. Union IT Minister Ashwini Vaishnaw announced the consultation on October 8 and said it would be published within a month. On October 23 he said the document would come out "in the coming weeks" and that the government would ask citizens, businesses and AI users for their views. He also suggested that the scale of new AI risks may justify legislation and further regulation. Neither report gave a release date.

A Shift From Guidelines

The consultation would mark a change in India's approach. In November 2025 the ministry published the India AI Governance Guidelines, which favoured existing law and voluntary measures over a new statute. The guidelines set out seven principles, including human oversight where feasible, accountability in proportion to potential harm, and transparency for users and regulators. They also proposed an AI Governance Group to coordinate policy across departments, supported by a Technology and Policy Expert Committee and an AI Safety Institute. The drafting committee found that some AI risks could be handled under current law. It flagged gaps in liability when AI causes harm, in how data protection rules apply to model development, and in copyright over training data.

Why It Matters

India is one of the largest user markets for global AI companies, and Indian ministers have made it a priority to attract AI investment. A move towards binding rules on safety, deepfakes and accountability would affect model providers, platforms and Indian businesses that deploy AI, including the many Gulf-based and multinational groups with large operations there. Companies operating in India should expect a comment window once the paper is published.